The Science of Shrinking Superconducting Qubits

Quantum technologies may be built on science that happens at extraordinarily small scales, but some of the most promising candidate platforms for quantum computers are nevertheless a bit big. To take advantage of the quantum mechanics that give quantum computers their powerful computing potential, their quantum bits, or qubits, must be protected from their environment; in the case of superconducting qubits, that means making them several millimeters large. That may still seem small, but hundreds if not thousands of qubits must be linked to yield a quantum computer that can best its conventional counterparts.

Shrinking superconducting qubits is the goal of SuperVan, a project sponsored by the Army Research Office and National Security Agency’s Laboratory for Physical Sciences (LPS) Qubit Collaboratory. This project is co-led by Columbia Engineering professor James Hone and BBN Technologies scientist Kin Chung Fong. The SuperVan team wants to create what’s known as a transmon qubit entirely from materials that can be as thin as a single atom.

In this Q&A, Columbia postdoc Jesse Balgley, who is co-advised by Hone and Fong and recently received a Quantum Computing Research Postdoctoral Fellowship from the ARO and LPS, explains more about the science of shrinking superconducting qubits and offers some words for aspiring quantum scientists and Columbians as a native New Yorker. 

What is a transmon qubit?

A transmon qubit is a quantum bit created from a macroscopic superconducting circuit, as opposed to, say, a subatomic quantum object like the spin of an electron. You use the electromagnetic modes of the circuit to create the 0 and 1 energy levels that serve to coherently perform computer operations.

How are you and the SuperVan team working to improve transmons? 

Current state-of-the-art transmons are typically grown from elemental superconductors whose surfaces form amorphous oxides. These are notorious sources of dissipation and decoherence, and such disorder can cause a qubit to lose its stored information quicker than operations can be performed on it. In the past two decades, most of the significant improvements to superconducting qubit coherence times have come from increasing the circuit footprint. But if you want to make a quantum processor that contains enough qubits to outperform a classical computer, you need to shrink the individual qubits.

Our idea is to instead use “cleaner” 2D crystalline materials, which can be isolated with very few defects and encapsulated in inert materials to protect them from forming lossy oxides. This will hopefully help us make a longer-lasting superconducting qubit with a footprint up to four orders of magnitude smaller than conventional transmons. 

The SuperVan team previously showed that the largest component in a transmon qubit, the shunt capacitor, can be replaced by a much smaller 2D-material-based capacitor and still show coherence. Now we are working towards constructing a transmon entirely out of 2D materials.

What will you be focusing on during the fellowship?

I’ll be exploring 2D materials as inductive elements for superconducting quantum devices. The fundamental building block of most superconducting qubits is called a Josephson junction, in which two superconductors sandwich another material—most commonly, an insulating barrier. The Josephson junction provides the necessary nonlinearity for the qubit circuit to endow it with addressable 0 and 1 states. It can also be placed in parallel with other circuit components to alter the qubit’s properties. When a Josephson junction is shunted with an inductor, a special type of qubit called a fluxonium is created, which has been shown to have dramatically enhanced dephasing times compared to transmon qubits. However, amorphous oxides and disorder are still typically prevalent in the materials used to make fluxonium qubits.

In the Hone lab, we grow high-purity crystals of 2D materials. We recently discovered that if you thin down a bulk crystal of a superconducting material called molybdenum ditelluride (MoTe₂) to a single atomic layer, its superconductivity becomes enhanced nearly one hundredfold!  As a superconductor gets thinner its intrinsic inductance increases—and you can’t get thinner than the atomic limit—so we expect to achieve exceptionally large inductances in flakes of MoTe₂. 

Our plan is to measure the magnitude of the inductance of these tiny flakes, which can be protected against oxidation, and then implement them in our 2D qubit circuits to make long-coherence-time fluxonium qubits with even smaller form factors than the state-of-the-art.

How does this work fit into your bigger-picture goals?

My research background is centered around 2D materials, and I was initially focused on their intrinsic electronic transport properties. In my current role, I am bridging 2D materials with superconducting qubits. 

Many research institutions, as well as major companies like IBM and Google, are putting a tremendous amount of effort and resources towards their quantum departments, with the ultimate goal of creating an operable quantum processor that can solve important problems, like developing new medicines or facilitating climate research in short order compared to what conventional computers can do. Thinking ahead, I’d like to be part of a team where I can continue the “bottom-up” approach of the SuperVan project, exploring new material platforms with the potential to disrupt the state of the art and advance quantum technology.

What sparked your initial interest in 2D materials?

2D materials research was really taking off when I was an undergraduate, and I was lucky to have had Professor Cory Dean as an advisor during the year he spent at City College. After I graduated, I helped with his transition to Columbia and spent a double gap year doing research in his lab. It was an extremely immersive introduction to the world of 2D materials, learning from some of the major innovators in this field in the very place where it was pioneered. 

I continued studying the electronic properties of 2D crystals during graduate school at Washington University in St. Louis, where I worked with Professor Erik Henriksen to explore charge transfer and magnetism in these materials. As I was finishing, Dr. Fong, who I briefly overlapped with at Columbia, reached out about a postdoc opportunity with the SuperVan team specifically to combine 2D materials with quantum computing.

It felt extremely serendipitous: the field I had been working in for almost ten years was being coupled with a rising field showing a lot of exciting possibilities and promise for technology. And it was a chance to come back to New York City. I was born and raised in Brooklyn. It’s home, and this city is my favorite place in the world. I feel very proud to be here at Columbia as part of the global push to get quantum technology off the ground. 

What have been some highlights from your time here?

I had countless mentors in those two years at Columbia before grad school. It was this special time when Professor Dean was moving into Philip Kim’s lab space as he was heading to Harvard, so many of his students and postdocs were still around. I got to work with so many people who have gone on to do incredible research, and I’m continually impressed when I read a new publication and find that a former colleague and mentor is the lead author.

Being in a place like Columbia, you are always going to be around people who have big ideas and are making things happen. I feel very fortunate to work with such motivated, collaborative people, and I strive to help guide the students I work with in the same way that others instructed me when I was starting out. I’m constantly thinking about the advice given to me by my mentors and, not only applying it to my work but also trying to share it with others. 

Do any particular words of advice stand out?

Working with 2D materials can be extremely difficult. They are atomically thin, extremely finicky, and often very fragile. You can try for weeks at something only to be met with so many instances of failure. My PhD advisor taught me that the best thing to do is take a step back, identify the roadblocks, and then determine if the process can be improved or if you should pivot and focus on something else instead. 

To continue to bang your head against the problem instead of gracefully bowing out is an example of the “sunk-cost fallacy” and can lead to burnout. What I try to impart to younger students is to not get lost in the weeds. Periodically take a bird’s-eye look at your project as a whole, list out all the failure modes, and see if these can be overcome. If not, don’t get discouraged because this kind of work is naturally very hard! You can always put your skills towards other projects and no doubt find success.

And what about advice about New York, as a native New Yorker?

In New York, it’s as simple as: go out and explore. Get outside of your box. There are so many opportunities to learn and find something that makes you happy. I truly believe you can do anything here.